JP7360379B2 - Image correction method and device compatible with perspective - Google Patents

Description

The present invention relates to a perspective-aware image correction method and an apparatus for carrying out the method, and in particular to a perspective-aware image correction method capable of providing improved legibility for a given target image. The present invention relates to an image correction method and apparatus capable of converting the target image into an image.

FIG. 1 shows an example of the use of conventional road markings.
Road markings refer to markings made with paint, paint, etc. on the road surface for the purpose of safe and smooth vehicle operation. The content varies from signs regarding vehicle operation standards, such as the central line and crosswalks, to descriptive signs such as ``slow speed'' and ``school zone.''

These road markings are now widely used on roads where vehicles are in operation, as they can provide drivers with a variety of information, such as driving information and warnings necessary for traffic safety. The reality is that

However, when a vehicle driver actually looks at road markings on a road inside the vehicle, perspective distortion may occur depending on the driver's viewing angle, the distance from the road markings, and the like. In particular, visual distortion may occur in which the width of the image, especially the vertical width, becomes narrower and more concentrated as the distance from the observation viewpoint increases. Therefore, there is a problem in that the legibility of road markings deteriorates due to such visual distortion.

In particular, in the case of road markings in South Korea, the structure of the character image is more complex than that of the alphabet due to the use of final consonants and hard-sounding characters. Therefore, a driver may be unable to read or misread road markings from a long distance due to visual distortions caused by perspective. Although such a phenomenon can be seen even with road markings consisting of one line, it may occur more frequently when road markings consisting of two or more lines are provided.

In order to solve this problem, in the prior art, it has been common practice to ensure legibility by simply extending the length of road markings along the direction in which the driver views them.

However, because the specific matters and extent of road marking adjustments are often determined based on the contractor's subjective judgment criteria and construction convenience, the effect of improving legibility is insufficient, and consistent improvements cannot be made. It was difficult to provide any effect.

In particular, in the conventional technology, the length of road markings was simply extended at the same ratio overall along the direction in which the driver views them, so when viewed from the driver's field of view, horizontal markings located far from the driver's field of vision There still existed the problem that the thickness of the image appeared to be thin, and the thickness of the horizontal image located near the field of view appeared to be thick. Therefore, it is difficult to instantly understand the meaning of road markings in a fast-moving car, which leads to problems that threaten safety.

These problems are not just problems with road markings displayed on roads. This is also a problem with Street View, which is often used on portal sites these days. Figure 3 is a screenshot of Google Street View. On the screen, the words ``Meadowlands Pkwy'' were added to the street view using a computer to display the street name on the road. However, as we added perspective to provide a more realistic feel to Street View users, as you can see in the drawing, the legibility deteriorates noticeably the further away you are from the viewpoint. I understand.

Referring to FIG. 2, this will be schematically explained as follows.

FIG. 2a shows the situation viewed from a viewpoint located at a certain height above the ground. In the drawings, lines marked HL indicate horizontal lines arranged parallel to each other at equal intervals, and lines marked VL also indicate vertical lines arranged parallel to each other at equal intervals. The line VL is orthogonal to each line HL.

However, due to perspective, if you look down at the ground from a distance, as shown in the figure, the vertical lines VL will appear to converge toward an infinitely located vanishing point, and the farther you are from the viewpoint, the greater the distance between adjacent vertical lines. appears to be narrowing. The distance between adjacent horizon lines HL also appears to become narrower as the distance from the viewpoint increases. Therefore, for example, if it is assumed that characters and symbols are arranged in the same size within the rectangle 10 in FIG. 2, it can be seen that legibility deteriorates significantly as the distance from the viewpoint increases.

Furthermore, in computer games (e.g. racing games), augmented reality (AR), virtual reality (VR), etc., legibility deteriorates when perspective is applied to express characters and symbols. A problem may arise.

FIG. 2b shows a space including four surfaces: a top surface, a bottom surface, and left and right side surfaces. In VR, AR, etc., graphics (signs) are not only added to the bottom surface as in the example of road markings mentioned above, but also to the left and right sides and the top surface. It can be seen from FIG. 2b that perspective distortion phenomena can also occur in this case, for example if characters are added to the left side.

Therefore, in composing various images displayed on a certain surface and providing predetermined information, such as road markings, it is possible to provide a consistent and easy image correction method, and to achieve harmony with the surrounding environment. What is needed is an image generation device and method of operation that can provide improved legibility.

Meanwhile, in order to solve the above problems, Korean Patent Publication No. 10-1668802 discloses a method of operating a long-distance identification image generation device.

Specifically, in the patent publication,
(a) providing a first image including at least one character;
(b) generating a second image in which the ratio of the first image is changed by reflecting viewpoint information on the first image and applying reverse perspective; (including information regarding at least one of the line of sight direction and viewing angle),
(c) extracting the coordinates of a first reference point of the first image and the coordinates of a second reference point of the second image that corresponds to the first reference point; (extracted based on at least one of the following),
(d) comparing at least one value of the coordinates of the first reference point and at least one value of the coordinates of the second reference point corresponding to the first reference point; transforming the first image by enlarging or reducing at least a portion of the reference region to be divided along the viewing direction;
(e) The step of generating the second image includes the step of converting the first image into a three-dimensional image, and the step of converting the first image converted into the three-dimensional image into a predetermined image by reflecting the viewpoint information. Disclosed is an operating method characterized by comprising the step of generating the second image by rotating the second image by a predetermined rotation angle in the viewing direction around the rotation axis of the second image and extracting a planar image.

However, the method involves converting the first image into a three-dimensional image, rotating it around the rotation axis, and then extracting the plane image again, and determining the reference point coordinates of the first image and the second image. It requires a process of extraction.

The present invention solves the above-mentioned problems and provides a consistent and easy image correction method without the need for rotating an image or extracting a separate reference point, and also provides improved legibility. Provided are an image correction method and apparatus capable of providing.

According to an embodiment of the present invention, an image correction method includes the steps of: providing a first image; providing viewpoint information for viewing the first image; dividing the first image into two or more divided images; and converting each of the two or more divided images based on the viewpoint information and the vertical length of each of the divided images; An image generation method is provided that includes providing a transformed second image.

In another embodiment, the viewpoint information includes information that allows the position of the viewpoint to be specified with respect to the first image.

In another embodiment, the viewpoint information includes a viewpoint height from a plane on which the first image is located to the viewpoint, and a height from the point where the viewpoint is projected onto the plane on which the first image is located. A viewpoint distance from the viewpoint side end of one image, a viewing angle at the viewpoint side end of the first image, a viewing angle at the opposite end of the viewpoint of the first image, and a viewpoint from the viewpoint to the first image. It includes two or more of the following: a line-of-sight length to a side edge, a line-of-sight length from the viewpoint to an end opposite to the viewpoint of the first image, and a viewpoint angle of the first image.

In another embodiment, the step of providing a second image transformed from the first image includes determining a field of view length for each divided image based on the viewpoint information and the vertical length of each of the divided images. and determining a transformation length for each divided image from the field of view length.

In another embodiment, the step of calculating the transform length for each divided image from the field of view length may be such that the ratio of the transform length of the i-th divided image to the total transform length is the total field of view length. The method includes the step of determining a transformation length for each divided image so that the ratio of the field of view length of the ni+1th divided image among the divided images is the same as that of the field of view length. Here, n is the total number of divided images.

In another embodiment, the step of calculating the transform length for each divided image from the field of view length is the step of calculating the transform length y _i of the i-th divided image using the following conversion formula Equation 1. including

ここで、ｈ_ｉはｉ番目分割イメージの視野長さ、ｎは分割イメージの総個数、Ｔは対象イメージの全体長さである。

Here, h _i is the visual field length of the i-th divided image, n is the total number of divided images, and T is the total length of the target image.

In another embodiment, the step of calculating the visual field length is a step of calculating the visual field length using a visual field length calculation formula, and the visual field length h _i for the i-th divided image is calculated as follows. The number is 2.

ここで、Ｓは視点から焦点までの視線長さであり、θ_ｉはｉ番目分割イメージの視線反対側端部での視野角であり、θ_０は１番目分割イメージの視線側端部での視野角であり、θ_ｆは焦点での視野角である、視野長さの算出式によって視野長さを求めるステップを含む。

Here, S is the line-of-sight length from the viewpoint to the focal point, θ _i is the viewing angle at the end of the i-th divided image on the opposite side of the line-of-sight, and θ ₀ is the viewing angle at the end of the first divided image on the side of the line-of-sight. The method includes the step of determining the visual field length using a visual field length calculation formula, where θ _f is the visual angle at the focal point.

In another embodiment, the step of calculating the visual field length is a step of calculating the visual field length using a visual field length calculation formula, and the visual field length h _i for the i-th divided image is calculated as follows. The number is 3.

Here, R is the line-of-sight length from the viewpoint to the end of the target image on the viewpoint side, θ _i is the viewing angle at the end of the i-th divided image opposite to the viewpoint, and θ ₀ is the length of the line of sight from the viewpoint to the end of the target image on the viewpoint side. The method includes the step of determining the visual field length using a visual field length calculation formula, which is the visual angle at the line-of-sight side end of the divided image.

In another embodiment, the step of providing a second image transformed from the first image includes determining a viewpoint angle for each divided image based on the viewpoint information and the vertical length of each of the divided images. and determining a transformation length for each divided image from the viewpoint angle.

In another embodiment, the step of determining the viewpoint angle for each divided image includes the step of determining the viewpoint angle for each divided image such that the ratio between the viewpoint angles for each divided image and the ratio between the vertical lengths of each divided image are the same. The method includes the step of determining a viewpoint angle for each divided image.

In another embodiment, the step of determining the viewpoint angle for each divided image includes the step of determining the viewpoint angle α _i for the i-th divided image using the following conversion equation (4).

ここで、Ｔは対象イメージの全体縦長さ、ｎは分割イメージの総個数、ａ_ｉはｉ番目分割イメージの縦長さ、θ_０は１番目分割イメージの視線側端部での視野角、θ_ｎはｎ番目分割イメージの視線反対側端部での視野角である。

Here, T is the overall vertical length of the target image, n is the total number of divided images, a _i is the vertical length of the i-th divided image, θ ₀ is the viewing angle at the line-of-sight side end of the first divided image, θ _n is the viewing angle at the end of the n-th divided image on the opposite side of the line of sight.

In another embodiment, the step of determining the transform length for each divided image includes the step of determining the transform length y _i of the i-th divided image using the following transformation equation (5).

ここで、Ｈは前記第１イメージが位置する平面から視点までの視点高さ、Ｌは前記視点を前記第１イメージが位置する平面上に投影させた地点から前記第１イメージの視点側端部までの視点距離である。

Here, H is the viewpoint height from the plane where the first image is located to the viewpoint, and L is the viewpoint side end of the first image from the point where the viewpoint is projected onto the plane where the first image is located. This is the viewing distance.

According to another embodiment of the invention, a computer readable memory includes computer readable instructions, the instructions, when executed on a computer, cause the computer to provide a first image. an operation of providing viewpoint information for viewing the first image; an operation of dividing the first image into two or more divided images along a lateral direction of the first image; and an operation of dividing the first image into two or more divided images, and the two or more divided images. and converting each of the divided images based on the viewpoint information and the vertical length of each of the divided images, the computer-readable method is configured to perform an act that includes providing a second image that is transformed from a first image. memory is provided.

In another embodiment, the viewpoint information includes information that allows the position of the viewpoint to be specified with respect to the first image.

In another embodiment, the viewpoint information includes a viewpoint height from a plane on which the first image is located to the viewpoint, and a height from the point where the viewpoint is projected onto the plane on which the first image is located. A viewpoint distance from the viewpoint side end of one image, a viewing angle at the viewpoint side end of the first image, a viewing angle at the opposite end of the viewpoint of the first image, and a viewpoint from the viewpoint to the first image. It includes two or more of the following: a line-of-sight length to a side edge, a line-of-sight length from the viewpoint to an end opposite to the viewpoint of the first image, and a viewpoint angle of the first image.

In another embodiment, the act of providing a second image transformed from the first image includes determining a field of view length for each divided image based on the viewpoint information and the vertical length of each of the divided images. and an operation of determining a transformation length for each divided image from the field of view length.

In another embodiment, the ratio of the transform length of the i-th divided image to the total transform length is the same as the ratio of the field-of-view length of the ni+1-th divided image to the total field-of-view length. As described above, the operation of determining the transform length for each divided image from the field of view length includes the operation of determining the transform length for each divided image. Here, n is the total number of divided images.

In another embodiment, the operation of calculating the transform length for each divided image from the field of view length is the operation of calculating the transform length y _i of the i-th divided image using the following conversion formula Equation 6. including.

ここで、ｈ_ｉはｉ番目分割イメージの視野長さ、ｎは分割イメージの総個数、Ｔは対象イメージの全体長さである。

Here, h _i is the visual field length of the i-th divided image, n is the total number of divided images, and T is the total length of the target image.

In another embodiment, the operation of calculating the field of view length is an operation of calculating the field of view length using a calculation formula for the field of view length, and the calculation formula for the field of view length h _i for the i-th divided image is as follows. The method includes the step of determining the visual field length using the visual field length calculation formula, which is Equation 7.

ここで、Ｓは視点から焦点までの視線長さであり、θ_ｉはｉ番目分割イメージの視線反対側端部での視野角であり、θ_０は１番目分割イメージの視線側端部での視野角であり、θ_ｆは焦点での視野角である。

Here, S is the line-of-sight length from the viewpoint to the focal point, θ _i is the viewing angle at the end of the i-th divided image on the opposite side of the line-of-sight, and θ ₀ is the viewing angle at the end of the first divided image on the side of the line-of-sight. is the viewing angle, and θ _f is the viewing angle at the focal point.

In another embodiment, the operation of calculating the field of view length is an operation of calculating the field of view length using a calculation formula for the field of view length, and the calculation formula for the field of view length h _i for the i-th divided image is as follows. This includes the operation of calculating the field of view length using the formula for calculating the field of view length, which is number 8.

ここで、Ｒは視点から対象イメージの視点側端部までの視線長さであり、θ_ｉはｉ番目分割イメージの視点反対側端部での視野角であり、θ_０は１番目分割イメージの視線側端部での視野角である。

Here, R is the line-of-sight length from the viewpoint to the end of the target image on the viewpoint side, θ _i is the viewing angle at the end opposite to the viewpoint of the i-th divided image, and θ ₀ is the viewing angle of the ith divided image at the opposite end of the viewpoint. This is the viewing angle at the line-of-sight side end.

In another embodiment, the act of providing a second image transformed from the first image determines a viewpoint angle for each divided image based on the viewpoint information and the vertical length of each of the divided images. and an operation of determining the transformation length for each divided image from the viewpoint angle.

In another embodiment, the operation of determining the viewpoint angle for each divided image is performed such that the ratio between the viewpoint angles for each divided image and the ratio between the vertical lengths of each divided image are the same. This includes the operation of determining the viewpoint angle for each divided image.

In another embodiment, the operation of determining the viewpoint angle for each divided image includes the operation of determining the viewpoint angle α _i for the i-th divided image using the following conversion equation (9).

ここで、Ｔは対象イメージの全体縦長さ、ｎは分割イメージの総個数、ａ_ｉはｉ番目分割イメージの縦長さ、θ_０は１番目分割イメージの視線側端部での視野角、θ_ｎはｎ番目分割イメージの視線反対側端部での視野角である。

Here, T is the overall vertical length of the target image, n is the total number of divided images, a _i is the vertical length of the i-th divided image, θ ₀ is the viewing angle at the line-of-sight side end of the first divided image, θ _n is the viewing angle at the end of the n-th divided image on the opposite side of the line of sight.

In another embodiment, the operation of determining the transform length for each divided image includes the operation of determining the transform length y _i of the i-th divided image using the following transformation equation (10).

ここで、Ｈは前記第１イメージが位置する平面から視点までの視点高さ、Ｌは前記視点を前記第１イメージが位置する平面上に投影させた地点から前記第１イメージの視点側端部までの視点距離である。

Here, H is the viewpoint height from the plane where the first image is located to the viewpoint, and L is the viewpoint side end of the first image from the point where the viewpoint is projected onto the plane where the first image is located. This is the viewing distance.

According to yet another embodiment of the present invention, the image correction device includes an input unit configured to input a first image and viewpoint information for viewing the first image, and a side view of the first image. dividing the first image into two or more divided images along a direction, and converting each of the two or more divided images based on the viewpoint information and the vertical length of each of the divided images; An image correction apparatus is provided that includes an image converter configured to convert a first image into a second image.

In another embodiment, the image correction device further includes an output unit configured to output the transformed second image.

In another embodiment, the viewpoint information includes information that allows the position of the viewpoint to be specified with respect to the first image.

In another embodiment, the viewpoint information includes a viewpoint height from a plane on which the first image is located to the viewpoint, and a height from the point where the viewpoint is projected onto the plane on which the first image is located. A viewpoint distance from the viewpoint side end of one image, a viewing angle at the viewpoint side end of the first image, a viewing angle at the opposite end of the viewpoint of the first image, and a viewpoint from the viewpoint to the first image. It includes two or more of the following: a line-of-sight length to a side edge, a line-of-sight length from the viewpoint to an end opposite to the viewpoint of the first image, and a viewpoint angle of the first image.

In another embodiment, the image conversion unit calculates a visual field length for each divided image based on the viewpoint information and the vertical length of each divided image, and calculates the visual field length for each divided image from the visual field length. The first image is configured to be transformed into a second image by determining a transformation length for the image.

In another embodiment, the image conversion unit is configured such that the ratio of the conversion length of the i-th divided image to the total conversion length is equal to the visual field length of the n−i+1st divided image of the total visual field length. The transform length for each divided image is determined such that the ratio occupied by the split image is the same as that of the split image. Here, n is the total number of divided images.

In another embodiment, the image transformation unit is configured to obtain the transformation length y _i of the i-th divided image using the following transformation equation (11).

ここで、ｈ_ｉはｉ番目分割イメージの視野長さ、ｎは分割イメージの総個数、Ｔは対象イメージの全体長さである。

Here, h _i is the visual field length of the i-th divided image, n is the total number of divided images, and T is the total length of the target image.

Ｓは視点から焦点までの視線長さであり、θ_ｉはｉ番目分割イメージの視線反対側端部での視野角であり、θ_０は１番目分割イメージの視線側端部での視野角であり、θ_ｆは焦点での視野角である。 In another embodiment, the image conversion unit is configured to calculate the visual field length using a visual field length calculation formula. Here, the formula for calculating the visual field length h _i for the i-th divided image is the following Equation 12.

S is the line-of-sight length from the viewpoint to the focal point, θ _i is the viewing angle at the end of the i-th divided image on the opposite side of the line-of-sight, and θ ₀ is the viewing angle at the end of the first divided image on the side of the line-of-sight. , and θ _f is the viewing angle at the focal point.

In another embodiment, the image conversion unit is configured to calculate the visual field length using a visual field length calculation formula. Here, the formula for calculating the visual field length h _i for the i-th divided image is the following equation 13.

Ｒは視点から対象イメージの視点側端部までの視線長さであり、θ_ｉはｉ番目分割イメージの視点反対側端部での視野角であり、θ_０は１番目分割イメージの視線側端部での視野角である。

R is the line-of-sight length from the viewpoint to the end of the target image on the view-side side, θ _i is the viewing angle at the end opposite to the viewpoint of the i-th divided image, and θ ₀ is the line-of-sight end of the first divided image This is the viewing angle in the area.

In another embodiment, the image conversion unit determines a viewpoint angle for each divided image based on the viewpoint information and the vertical length of each of the divided images, and calculates a viewpoint angle for each divided image from the viewpoint angle. The first image is configured to be transformed into a second image by determining a transform length.

In another embodiment, the image conversion unit converts each divided image so that a ratio between viewpoint angles and a ratio between vertical lengths of each divided image are the same. It is configured to find the viewing angle.

In another embodiment, the image conversion unit is configured to obtain the viewpoint angle α _i for the i-th divided image using the following conversion equation (14).

ここで、Ｔは対象イメージの全体縦長さ、ｎは分割イメージの総個数、ａ_ｉはｉ番目分割イメージの縦長さ、θ_０は１番目分割イメージの視線側端部での視野角、θ_ｎはｎ番目分割イメージの視線反対側端部での視野角である。

Here, T is the overall vertical length of the target image, n is the total number of divided images, a _i is the vertical length of the i-th divided image, θ ₀ is the viewing angle at the line-of-sight side end of the first divided image, θ _n is the viewing angle at the end of the n-th divided image on the opposite side of the line of sight.

In another embodiment, the image transformation unit is configured to obtain the transformation length y _i of the i-th divided image using the following transformation equation (15).

Here, H is the viewpoint height from the plane where the first image is located to the viewpoint, and L is the viewpoint side end of the first image from the point where the viewpoint is projected onto the plane where the first image is located. This is the viewing distance.

With the image correction method and apparatus provided by the present invention, it is possible to improve the phenomenon in which legibility deteriorates due to the effect of perspective even when viewing an image from a long distance.

Therefore, when the present invention is applied to road markings, it is possible for a driver to read the road markings from a further distance. This means that the reaction time can be reduced by the difference in distance, thus allowing the driver to make quick decisions even when traveling at high speed. Furthermore, it is possible to prevent traffic accidents due to unreasonable interruptions that may occur due to the phenomenon of missing the exit at a junction or the like or delayed interpretation.

Furthermore, if applied to VR or AR environments, it will be possible to realize images that give a sense of perspective while improving legibility. Therefore, a more comfortable VR or AR environment can be constructed.

Other effects of the present invention will be easily understood by those with ordinary knowledge in the technical field to which the present invention pertains from the content described in the specification of the present invention.

An example of the use of conventional road markings is shown.

Shows the state viewed from a viewpoint located at a certain height above the ground.

Shows how the space is viewed from a spatial perspective.

This is a screenshot of Google Street View.

1 schematically illustrates an image correction method according to an embodiment of the invention.

5 schematically shows an embodiment modified with respect to the embodiment of FIG. 4;

3 schematically illustrates an image correction method according to another embodiment of the invention.

7 schematically shows an embodiment modified with respect to the embodiment of FIG. 6;

5 schematically illustrates an image correction method according to another embodiment of the present invention.

4 shows a target image and a deformed image transformed according to the present invention.

The target image and the deformed image in FIG. 9a are shown viewed from a fixed viewpoint.

4 shows a target image and a deformed image transformed according to the present invention.

The target image and the deformed image in FIG. 10a are shown viewed from a fixed viewpoint.

A target image added to the left side in space and a deformed image deformed according to the present invention are shown as viewed from a certain viewpoint.

4 shows the street view capture screen of FIG. 3 (top view) and an image converted from it according to the present invention (bottom view).

1 schematically shows the configuration of an image correction device according to an embodiment of the present invention.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Image generation apparatuses and methods according to embodiments of the present invention will be described below with reference to the accompanying drawings.

It should be made clear that the following embodiments are detailed explanations to help understand the present invention, and are not intended to limit the scope of the present invention. Therefore, equivalent inventions that perform the same functions as the present invention would also fall within the scope of the present invention.

Also, when assigning reference numerals to the constituent elements of each drawing, it should be noted that the same constituent elements are given the same reference numerals as much as possible when they appear on other drawings. There must be. In addition, in the description of the present invention, if it is determined that specific description of related publicly known configurations or functions may unnecessarily obscure the gist of the present invention, such detailed description shall not be provided. Omitted.

Further, in describing the constituent elements of the present invention, terms such as first, second, A, B, (a), (b), etc. can be used. These terms are used only to distinguish the component from other components, and do not limit the nature, procedure, or order of the component. When a component is referred to as being "coupled" or "connected" to another component, it may be directly coupled or connected to the other component; It is to be understood that further other components may be present in between. On the other hand, when a component is referred to as being "directly coupled" or "directly connected" to another component, it is assumed that there are no additional components in between. Must be understood. Other expressions that describe relationships between components, such as "between" and "immediately between" or "adjacent to" and "directly adjacent to," are interpreted similarly. Must.

Additionally, the terms used in this application are merely used to describe particular embodiments and are not intended to limit the invention. Singular expressions include plural expressions unless the context clearly dictates otherwise. In this application, terms such as "comprising" or "having" are intended to specify the presence of features, numbers, steps, acts, components, parts, or combinations thereof that are described in the specification. It is to be understood that this does not exclude in advance the existence or possibility of addition of one or more other features, figures, steps, operations, components, parts or combinations thereof. .

Unless otherwise defined, all terms used herein, including technical and scientific terms, have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. has. Terms as defined in commonly used dictionaries shall be construed to have meanings consistent with the meanings they have in the context of the relevant art, and unless expressly defined herein, ideal or It should not be interpreted as an overly formal meaning.

In the present invention, the target image refers to the original image (first image) given before being transformed by the method of the present invention. Such target images are, for example, symbols and characters such as road markings (school zone, stop, slow down, etc.) displayed on the road surface, or road surfaces inserted on a computer such as Google Street View. This includes signs, characters and symbols inserted in virtual reality (VR) or augmented reality (AR), and other images realized in reality or on computer graphics.

In the present invention, a modified image refers to an image (second image) obtained by transforming a given target image using the method presented by the present invention.

In the present invention, the viewpoint side edge refers to a rectangle whose boundaries are the top, bottom, leftmost, and rightmost parts of a given image. The point closest to the viewpoint among the nearby sides). In the present invention, the end opposite to the viewpoint refers to the point on the upper side of the rectangle that is closest to the viewpoint. For example, in FIG. 4, point A is the end of the target image (or the first divided image) on the viewpoint side, and point E is the end of the target image (or the fourth divided image) on the opposite side of the viewpoint.

In the present invention, the horizontal direction and the vertical direction refer to directions viewed from the rectangle. For example, in Figures 2a and 2b, HL indicates a horizontal line and VL indicates a vertical line.

In the present invention, ground surface refers to any surface on which the target image is applied or added by computer graphics, including but not limited to the bottom surface, side surfaces, and ceiling surfaces. It's okay. For example, when a graphic is added to the left side as shown in FIG. 2b, the left side also corresponds to the ground.

In the present invention, the line-of-sight length refers to the distance from the viewpoint to the relevant point. For example, in the embodiment described below, this applies to the length of line segment OA, the length of line segment OE, etc.

In the present invention, the field of view length refers to the vertical length of a given image from the end on the viewpoint side to the end on the opposite side from the viewpoint as observed from the viewpoint. That is, it is the vertical length of a given image from the end on the viewpoint side to the end on the opposite side of the viewpoint, which appears smaller than the actual length due to the application of perspective when viewed from the viewpoint. For example, in the following embodiments, this corresponds to the length of line segment M ₁ M ₂ or h _i .

In the present invention, the viewing angle refers to the angle at which a viewpoint is viewed from one point on the ground. For example, in the embodiment described below, the viewing angle at point A is θ ₀ and the viewing angle at point C is θ ₂ .

In the present invention, the viewpoint angle refers to an angle formed when viewing a specific image from a viewpoint from an end on the viewpoint side to an end on the opposite side to the viewpoint. For example, in the embodiment described below, the viewpoint angle for the divided image 201 is ∠AOB, and the viewpoint angle for the entire target image 200 is ∠AOE.

In the present invention, viewpoint information refers to information sufficient to specify the position of a viewpoint with respect to a target image. For example, in FIG. 4, the viewpoint information includes the viewpoint distance L, the viewpoint height H, the viewing angle at the end of the target image on the viewpoint side, the viewing angle at the end of the target image opposite to the viewpoint, and the distance to the end of the target image on the viewpoint side. If at least two of the line-of-sight length, the line-of-sight length to the opposite end of the target image viewpoint, and the viewpoint angle with respect to the target image are known, the relative position of the viewpoint with respect to the target image can be determined.

In the present invention, when indicating an order such as the i-th, j-th, etc., the first, second, third, . ．． ．． We will decide the order as follows.

In the present invention, when a target image is given that is to be added physically directly on the ground such as a road or virtually to be added using computer graphics etc. so that perspective can be felt, the target image is We propose a method for converting the target image to solve the problem of poor legibility.

In a specific embodiment of the present invention, a given target image is divided into a fixed number of images along the horizontal direction, and among each of the divided images, a divided image located near the viewpoint is selected. The vertical length of divided images located far from the viewpoint is relatively increased. By converting in this manner, it is possible to maintain the overall vertical length of the target image that was originally intended to be added, while improving the problem of poor legibility due to perspective.

In another specific embodiment of the present invention, a given target image is horizontally divided into a fixed number of images, and then the field of view length of each divided image is determined. Thereafter, the ratio between these field of view lengths is applied inversely to each divided image to determine the vertical length (hereinafter referred to as conversion length) to which each divided image is transformed. In other words, when a given target image is divided into n divided images, the ratio of the transformation length of the i-th divided image to the total transformation length is equal to the ratio of the transformation length of the ith divided image to the entire field of view length. Find the transformation length so that it is the same as the ratio occupied by the field of view length.

FIG. 4 is a diagram for explaining a method according to an embodiment of the present invention. FIG. 4(b) is a plan view of the position where the target image 200 is placed, and FIG. 4(a) schematically shows a cross section taken along line GE in FIG. 4(b).

In FIG. 4, reference numeral 100 indicates a ground surface, and O indicates a view point. The viewpoint may correspond to, for example, the eyes of a driver riding in a car, but is not limited thereto. G indicates a point on the ground 100 located vertically downward from the viewpoint O. The height of the viewpoint O (viewpoint height), that is, the length of the line segment OG is indicated by H. The distance (viewpoint distance) from the point where the viewpoint O is projected on the ground, ie, point G, to the end of the target image 200 on the viewpoint side, ie, point A, is indicated by L.

Reference numeral 200 indicates a rectangle bounded by the top, bottom, leftmost, and rightmost parts of the target image (eg, road marking) to be added to the ground 100. 201, 202, 203, and 204, which will be described later, similarly indicate rectangles for each divided image. However, in the following, for convenience of explanation, the target image 200, divided images 201, etc. will be simply displayed. Additionally, although the target image 200 is shown as having a thickness below the ground 100 in the drawings, this is only for convenience of explanation; in reality, it may have no substantial thickness. , is assumed to be placed alongside the ground 100 on an extension of the ground 100.

In this embodiment, the target image 200 is divided into four divided images 201, 202, 203, and 204 in parallel in the horizontal direction (X-axis direction). Points A, B, C, D, and E are points where the horizontal line of the image divided in this manner passes through point G and intersects a line parallel to the Y axis. The vertical lengths (lengths in the Y-axis direction) of the divided images 201, 202, 203, and 204 were displayed as a ₁ , a ₂ , a ₃ , and a ₄ , respectively.

Although the target image 200 is divided into four parts in this embodiment and the following embodiments, it should be made clear that this is for convenience of explanation. A person having ordinary knowledge in the technical field to which the present invention pertains will understand from the content disclosed in the specification of the present invention that the more the number of parts into which the target image 200 is divided, the better the legibility becomes. can be expected to bring.

On the other hand, in this embodiment, it is assumed that the focal point is located on point C. The focal plane past the focal point is designated by the drawing numeral 300 and is orthogonal to the line of sight OC at point C. Points where the focal plane 300 intersects the line of sight OA, line of sight OB, line of sight OC, line of sight OD, and line of sight OE are indicated by M ₁ , M ₂ , M ₃ , M ₄ , and M ₅ , respectively. Therefore, in this embodiment, point C and point _M3 indicate the same point.

Here, the length of the divided image 201 observed from the viewpoint O, that is, the visual field length, is equal to the length of the line segment M ₁ M ₂ . Similarly, the field of view lengths of the divided image 202, the divided image 203, and the divided image 204 are the length of the line segment M ₂ M ₃ , the length of the line segment M ₃ M ₄ , and the length of the line segment M ₄ M _5, respectively. is equal to

If the line-of-sight length from the viewpoint to the focal point is S, it is equal to the length of the line segment OC, and therefore can be determined from the right triangle OCG as follows.

(1)

Furthermore, the viewing angles θ ₀ , θ ₁ , θ ₂ , θ ₃ , and θ ₄ can be determined as follows.

(2)

(3)


(4)

(5)

(6)

On the other hand, ∠AOB, ∠BOC, ∠COD, and ∠DOE are (θ ₀ −θ ₁ ), (θ ₁ −θ ₂ ), (θ ₂ −θ ₃ ), and (θ ₃ −θ ₄ ), respectively. .

Therefore, the field of view length of the divided images 201, 202, 203, 204, that is, the length of line segment M ₁ M ₂ , the length of line segment M ₂ M ₃ , the length of line segment M ₃ M ₄ , the line segment If the lengths of M ₄ M ₅ are h ₁ , h ₂ , h ₃ , and h ₄ respectively, these can be calculated as follows.

(7)
(8)
(9)
(10)

Further, if the lengths to which each divided image is converted are respectively y ₁ , y ₂ , y ₃ , and y ₄ , then it can be determined as follows.

(11)
(12)
(13)
(14)

Here, T represents the overall length of the target image 200 to be applied to the ground 100 or added by computer graphics work (for example, T=a ₁ +a ₂ +a ₃ +a ₄ in this embodiment).

Although the embodiment of FIG. 4 has been described with reference to the case where the target image 200 is divided into four pieces, the above formula can also be applied to the case where the target image 200 is divided into four or more pieces.

When dividing the target image 200 into n divided images, the line-of-sight length S from the viewpoint to the focal point, and the field-of-view length _{h i} of the i-th divided image starting from the divided image located closest to the viewpoint (first divided image) The viewing angle θ _i of the i-th divided image viewed from the end opposite to the viewpoint can be determined as follows.

(15)
(16)
(17)

Therefore, the lengths to which each segmented image is transformed are y ₁ , y ₂ , y ₃ , . ．． ．． When y _n , the converted length of the i-th divided image can be found as follows:

(18)

The table below shows T=5m, L=15m, H=1.2m, n=10, and a ₁ =a ₂ =. ．． ．． It is a table showing θ _i , h _i , and y _i values for the case of =a ₁₀ =0.5 m.

As can be seen from the table, the deformed length y ₁₀ of the 10th divided image is approximately 1.7 times the deformed length y ₁ of the 1st divided image.

9 and 10 are diagrams showing a target image and a deformed image obtained using the y _i values obtained from Table 1 above.

In FIG. 9a, the left side is the target image "school zone", and the right side is the image transformed using the y _i values obtained from Table 1. In the target image, the vertical lengths of ``school'' and ``zone'' were the same, but in the transformed image, it can be seen that ``school'' became longer than ``zone.'' It can also be confirmed that there are changes in each horizontal stroke. For example, the horizontal stroke at the bottom of "Zone" is thinner in the deformed image, and the horizontal stroke at the top of "School" is thicker in the deformed image.

FIG. 9b shows how the target image and the deformed image are viewed from a distance. Both images have an overall length T of 5 m, and are viewed from a viewing distance L = 15 m and a viewing height H = 1.2 m. The left side is the target image, and the right side is the deformed image. When viewed from a distance, the "school" appears smaller than the "zone" in the target image, but it can be seen that the "school" and "zone" appear to be approximately the same size in the deformed image. Further, it can be seen that the thickness of the horizontal stroke of the target image appears to be thinner as the distance from the viewpoint increases, but the thickness of the horizontal stroke of the deformed image appears to be substantially the same throughout.

FIGS. 10a and 10b show a target image "PITTS-BURGH" and a modified image thereof, respectively. Since the explanation regarding each is the same as that of FIGS. 9a and 9b, duplicate explanation thereof will be omitted.

FIG. 11 shows a state in which the target image and the deformed image are added to the left side in space, instead of images being placed on the bottom surface as in FIGS. 9 and 10. This arrangement of images can mainly occur in AR or VR environments, but is not limited to this. For example, it can also be applied to billboards and road markings erected on the side of the road. Probably.

The upper drawing in FIG. 11 shows how the target image appears after perspective is applied, and the lower drawing shows how the deformed image looks after perspective was applied. As can be seen from the diagram, the characters in the target image become less visible as the distance from the viewpoint increases, but it can be seen that the deformed image appears clearly regardless of the viewpoint. In particular, if you look at the "Pkwy" part at the back, you can see that the transformed image appears to be expanded about twice as much.

FIG. 12 shows both the street view capture screen (top view) of FIG. 3 and a screen (bottom view) in which road markings are converted according to the present invention. As can be seen in the drawing, the legibility of the Pkwy portion was low before the conversion, but it was significantly improved after the conversion.

On the other hand, although FIG. 4 shows a case where the focal point is located at the center of the target image, the present invention can also be applied when the focal point is located at another point.

FIG. 5 shows one embodiment in this regard.

Similar to FIG. 4, FIG. 5(b) is a plan view of the position where the target image 200 is placed, and FIG. 5(a) schematically shows a cross section taken along line GE in FIG. 5(b).

In FIG. 5, the focus is located between the first divided image and the second divided image. Thus, the focal plane 300 is orthogonal to the line of sight OB at point B.

The line-of-sight length S from the viewpoint to the focal point is equal to the length of the line segment OB, and can therefore be determined from the right triangle OBG as follows.

(19)

If the field of view lengths of the divided images 201, 202, 203, and 204 are respectively h ₁ , h ₂ , h ₃ , and h ₄ , these can be calculated as follows.

(20)
(21)
(22)
(23)

The lengths y ₁ , y ₂ , y ₃ , and y ₄ to which each divided image is transformed can be determined using the above equations (11) to (14).

Similarly, even when the target image is divided into n images, the above equations (15) to (18) can be applied in the same or similar manner as follows.

(15)
(16')
(17')
(18)

In the embodiment shown in FIG. 4 and the embodiment shown in _FIG . can be generalized to.

(24)
(25)
(However, f is an integer from 0 to n)

After all, it can be seen that the same transformation length can be obtained if the viewpoint conditions (eg, θ ₀ , H, or L) are the same, regardless of which divided image the focal point is located.

FIG. 6 shows another alternative embodiment of the invention.

In this embodiment, instead of finding the field of view length on the focal plane, an approach method is used to find the field of view length for each divided image at a point located at the same distance from the viewpoint O.

Similar to the above, FIG. 6(b) is a plan view of the position where the target image 200 is placed, and FIG. 6(a) schematically shows a cross section taken along line GE in FIG. 6(b). .

In FIG. 6, the target image is divided into four sections AB, BC, CD, and DE, similar to the embodiment described above. Point _M1 is the point where a circle (hereinafter referred to as "circle O") centered on the viewpoint O and whose radius is the sight line length R from the viewpoint to the end of the target image on the viewpoint side intersects the sight line OA. be. Point _M2 is the point where the tangent drawn from point _M1 intersects the line of sight OB. Point _M3 is a point where a tangent drawn from the point where circle O and line of sight OB intersect with line of sight OC. Point _M4 is a point where a tangent drawn from the point where circle O and line segment OC touch line of sight OD. Point _M5 is a point where a tangent drawn from the point where circle O and line of sight OD intersect with line of sight OE.

It can be said that the visual field length when viewing the first divided image 201 from the viewpoint O is the length of the line segment M ₁ M ₂ .

If the visual field length when viewing the second divided image 202 from the viewpoint O is measured at the same position where the visual field length was measured for the first divided image 201, then the visual field length is a line. The length will be M ₂ M ₃ minutes.

Similarly, the length of the field of view when viewing the third divided image 203 and the length of the field of view when viewing the fourth divided image 204 from the viewpoint O are the length of the line segment M ₃ M ₄ and the length of the line segment M ₄ M, respectively. It will be ₅ long.

On the other hand, the length R of the radius can be determined from the right triangle OAG as follows.

(26)

The viewing angles θ ₀ , θ ₁ , θ ₂ , θ ₃ , and θ ₄ are as shown in equations (2) to (6). Similarly, ∠AOB, ∠BOC, ∠COD, and ∠DOE are (θ ₀ −θ ₁ ), (θ ₁ −θ ₂ ), (θ ₂ −θ ₃ ), and (θ ₃ −θ ₄ ), respectively. Become.

Therefore, the field of view length of the divided images 201, 202, 203, 204, that is, the length of line segment M ₁ M ₂ , the length of line segment M ₂ M ₃ , the length of line segment M ₃ M ₄ , the line segment If the lengths of M ₄ M ₅ are h ₁ , h ₂ , h ₃ , and h ₄ , respectively, these can be determined as follows.

(27)
(28)
(29)
(30)

The lengths y ₁ , y ₂ , y ₃ , and y ₄ to which each divided image is transformed can be determined using the above equations (11) to (14), respectively.

Although the embodiment shown in FIG. 6 is a case where the target image 200 is divided into four parts, the above formula can also be applied when the target image 200 is divided into four or more parts.

When the target image 200 is divided into n divided images, starting from the divided image located closest to the viewpoint (first divided image), the viewing angle θ _i as seen from the end opposite to the viewpoint of the i-th divided image is as follows. It can be determined using equation (15), and the field of view length h _i of the i-th divided image can be determined as follows.

(31)

Similar to what was described above, the lengths to which each segmented image is transformed are y ₁ , y ₂ , y ₃ , . ．． ．． When y _n , the length of the i-th divided image to be converted can be obtained using the above equation (18).

The table below shows T=5m, L=15m, H=1.2m, n=10, and a ₁ =a ₂ =. ．． ．． It is a table showing θ _i , h _i , and y _i values for the case of =a ₁₀ =0.5 m.
It can be seen that in the case of this embodiment as well, deformed length values almost similar to those in Table 1 appear.

FIG. 7 shows an alternative embodiment to the embodiment of FIG.

Similar to the above, FIG. 7(b) is a plan view of the position where the target image 200 is placed, and FIG. 7(a) schematically shows a cross section taken along line GE in FIG. 7(b). .

In FIG. 7, the target image is divided into four sections AB, BC, CD, and DE, similar to the embodiment described above. Point _M1 is the point where a line extending perpendicularly from point A to line of sight OA intersects line of sight OA, so A and _M1 indicate the same point. Point M ₂ is a point where a line extending perpendicularly from point M ₁ to line of sight OA intersects line of sight OB. Point M ₃ is a point where a line extending perpendicularly from point M ₂ to line of sight OB intersects line of sight OC. Point M ₄ is a point where a line extending perpendicularly from point M ₃ to line of sight OC intersects line of sight OD. Point _M5 is a point where a line extending perpendicularly to the line of sight OD from point _M4 intersects the line of sight OE.

The visual field length when viewing the first divided image 201 from the viewpoint O is the length of the line segment M ₁ M ₂ .

Although the distance from viewpoint O to point _M2 is larger than the distance from viewpoint O to point _M1 , the degree of the difference is at a negligible level. Therefore, the length of the line segment M ₂ M ₃ can be approximated to the visual field length for the second divided image 202 from the viewpoint O.

Similarly, the length of the line segment M ₃ M ₄ and the length of the line segment M ₄ M ₅ are approximated to the visual field length when viewing the third divided image 203 and the fourth divided image 204 from the viewpoint O. I can do it.

Also, ∠AOB, ∠BOC, ∠COD, and ∠DOE become (θ ₀ - _{θ 1} ), (θ ₁ - θ ₂ ), (θ ₂ - _{θ 3} ), and (θ ₃ - θ ₄ ), respectively. .

Here, if the lengths of line segments OM ₁ , OM ₂ , OM ₃ , and OM ₄ are respectively S ₁ , S ₂ , S ₃ , and S ₄ , then the lengths of S ₁ , S ₂ , S ₃ , and S ₄ are can be calculated as follows.

(32)
(33)
(34)
(35)

The viewing angles θ ₀ , θ ₁ , θ ₂ , θ ₃ , and θ ₄ are as shown in equations (2) to (6).

Therefore, the field of view length of the divided images 201, 202, 203, 204, that is, the length of line segment M ₁ M ₂ , the length of line segment M ₂ M ₃ , the length of line segment M ₃ M ₄ , the line segment If the lengths of M ₄ M ₅ are h ₁ , h ₂ , h ₃ , and h ₄ , respectively, these can be determined as follows.

(36)
(37)
(38)
(39)

The lengths y ₁ , y ₂ , y ₃ , and y ₄ to which each divided image is transformed can be determined using the above equations (11) to (14), respectively.

Although the embodiment shown in FIG. 7 is a case where the target image is divided into four parts, the above formula can also be applied to the case where the target image is divided into four or more parts.

When dividing the target image 200 into n divided images, starting from the divided image located closest to the viewpoint (first divided image), the viewing angle θ _i as seen from the end opposite to the viewpoint of the i-th divided image is determined by the above formula. (15), and the visual field length h _i of the i-th divided image can be determined as follows.

(40)

Here, S _i is as follows.

however,

) (41)

Similar to what was described above, the lengths to which each segmented image is transformed are y ₁ , y ₂ , y ₃ , . ．． ．． When y _n , the length to which the i-th divided image is transformed can be found using equation (18).

The table below shows T=5m, L=15m, H=1.2m, n=10, and a ₁ =a ₂ =. ．． ．． It is a table showing θ _i , h _i , and y _i values for the case of =a ₁₀ =0.5 m.

It can be seen that deformed length values similar to those in Tables 1 and 2 can be obtained in this embodiment as well.

FIG. 8 shows yet another alternative embodiment of the invention.

Similar to the above, FIG. 8(b) is a plan view of the position where the target image 200 is placed, and FIG. 8(a) schematically shows a cross section taken along line GE in FIG. 8(b). .

In FIG. 8, the viewpoint angle ∠AOE for the entire target image 200 from the viewpoint O is divided into the same number of parts as the target image 200 is to be divided, and the length to be transformed is calculated using the divided angles. Use an approach method that seeks for clarity.

In FIG. 8, the target image 200 is divided into four sections AB, BC, CD, and DE, similar to the embodiment described above. The vertical lengths of each of the divided images 201, 202, 203, and 204 were expressed as a ₁ , a ₂ , a ₃ , and a ₄ , respectively (see FIG. 8(b)). Points A, B, C, D, and E are points where the horizontal lines of the thus divided images 201, 202, 203, and 204 pass through point G and intersect with a line parallel to the Y axis.

In accordance with the number of divided images, the angle (∠AOE) at which the entire target image 200 is viewed from viewpoint O is also divided into four angles α ₁ , α ₂ , α ₃ , and α ₄ . At this time, each angle is divided into sizes that satisfy the following formula.

(42)

Points A, B', C', D', and E indicate the points where the lines dividing the angle intersect with the target image 200, respectively.

In this way, the target image is newly divided into four sections: AB' section, B'C' section, C'D' section, and D'E section.

Assuming a circle O whose center is point O and whose radius length is line segment OA, (if the radius length is sufficiently large) arc M ₁ M ₂ , arc M ₂ M ₃ , arc M ₃ M _4. The length of the arc M ₄ M ₅ may be substantially the same as the length of the AB' section, B'C' section, C'D' section, and D'E section observed from the viewpoint O. Therefore, the length of line segment AB', the length of line segment B'C', the length of line segment C'D', and the length of line segment D'E are the lengths to which the divided images are converted. It is determined as y ₁ , y ₂ , y ₃ , y ₄ .

When the overall vertical length of the target image 200 is T (T=a ₁ +a ₂ +a ₃ +a ₄ in this embodiment), the viewing angles θ ₀ and θ ₄ are as follows.

(43)
(44)

On the other hand, the ratio of the vertical length of the first divided image 201 to the entire vertical length of the target image 200 is a ₁ /T. Therefore, the viewpoint angle α ₁ for the divided image 201 can be determined as follows by considering the above equation (42).

(45)

Similarly, the viewpoint angles α ₂ , α ₃ , and α ₄ for the second to fourth divided images (202 to 204) can be determined as follows.

(46)
(47)
(48)

Now, the lengths y ₁ , y ₂ , y ₃ , and y ₄ to which the divided images are converted are calculated as follows.

(49)
(50)
(51)
(52)

Although the embodiment shown in FIG. 8 is a case where the target image 200 is divided into four parts, the above formula can also be applied when the target image 200 is divided into four or more parts.

When dividing the target image 200 into n divided images, starting from the divided image located closest to the viewpoint (first divided image), the transformation length y _i and viewpoint angle α _i of the i-th divided image are as follows. You can ask for it.

(53)
(54)

Here, θ ₀ and θ _n can be determined using the above formula (2) and formula (15), respectively.

The table below shows T=5m, L=15m, H=1.2m, n=10, and a ₁ =a ₂ =a ₃ =. ．． ．． It is a table showing θ _i and y _i values for the case where = a ₁₀ =0.5 m.

It can be seen that deformed length values similar to those in Tables 1 to 3 can be obtained in this embodiment as well.

As a further embodiment of the present invention, the above equations (18) and (54) can be transformed using the correction factor f _i as follows.

(55)
(56)

The correction factor f _i can take an appropriate value depending on the situation or conditions under which the image is applied or applied. For example, when applying Equation 40 below, the closer a divided image is to the viewpoint, the less it is converted, and the farther away from the viewpoint, the more it is converted.

The f _i value is not limited to this, and a person with ordinary knowledge in the technical field to which the present invention pertains will be able to take different appropriate values depending on the user's needs and surrounding conditions such as ground conditions. You will be able to know what will happen.

An embodiment of the present invention provides an image correction device that can implement the image correction method as described above.

FIG. 13 is a block diagram showing an implementation example of such an image correction device. Referring to FIG. 13, an image correction apparatus 1000 according to an embodiment of the present invention includes an input unit 1010, an image conversion unit 1020, and an output unit 1030.

The user can input information regarding the target image 200 to be transformed via the input unit 1010. The target image information may include an image file of the target image, a width and a length to which the target image is added (length T in the above embodiment), and the like.

Further, the user can input the number of parts into which the target image 200 is to be divided and the vertical length of each divided image (a ₁ , a ₂ , a ₃ , . . . ). Preferably, each segmented image can be segmented into the same vertical length (ie, a ₁ =a ₂ =a ₃ =...=a _n ).

Additionally, the user can input viewpoint information via the input unit 1010. Viewpoint information means information that can specify the position of a viewpoint with respect to the target image 200. For example, the viewpoint may be a distance away from the target image, a viewpoint angle, etc.

More specifically, for example, the viewpoint height H from the plane where the target image is located to the viewpoint, and the viewpoint side end of the target image from the point where the viewpoint is projected onto the ground 100 where the target image is located. a viewing angle at the end of the target image on the viewpoint side, a viewing angle at the end of the target image opposite to the viewpoint, a line-of-sight length from the viewpoint to the end of the target image on the viewpoint side, It may be a line-of-sight length from the viewpoint to an end of the target image opposite to the viewpoint, a viewpoint angle of the target image, or the like. All of the viewpoint information is not necessary because it is enough to specify the position of the viewpoint with respect to the target image 200. That is, for example, in the embodiment shown in FIG. 6, if only the viewpoint distance L and the viewpoint height H are given, the viewing angle θ ₀ and the sight line length R can be obtained from the technology to which the present invention belongs. It can be understood by anyone with ordinary knowledge in the field.

A target image input through the input unit 1010 is converted through the image conversion unit 1020 through the process described above. For example, by calculating the y _i value for the i-th divided image using the above formula (18) or formula (54), and increasing or decreasing the vertical length of the divided image according to the y _i value obtained in this way, the target Transform the image 200. As an example, in the embodiment of FIG. 4, if the vertical length of the divided image 201 is 1 m and the calculated _y1 value for this is 0.7 m, then the horizontal length of the divided image 201 is maintained. While doing so, it is reduced and deformed so that the vertical length becomes 0.7 m.

The output unit 1030 can output the image converted by the image conversion unit 1020 in a desired format. For example, the output unit 1030 can output an image converted into a file format such as jpg, TIFF, or XML, but this is just an example, and it may be output in any known format or format if necessary. good. In particular, when the image correction device according to the present invention is used to realize AR, VR, etc., it can be provided in a format required by the AR device or VR device.

An embodiment of the present invention may be implemented in the form of a storage medium, ie, a computer-readable medium, containing computer-executable instructions such as program modules executed by a computer. Such a computer-readable medium can record a program for executing the image conversion method described above.

Computer-readable media can be any available media that can be accessed by a computer and includes both volatile and nonvolatile media, separable and non-separable media. Computer-readable media can also include both computer storage media and communication media.

Computer storage media can be volatile and non-volatile, separable and non-separable, implemented in any method or technology for storage of information such as computer-readable instructions, data structures, program modules or other data. Contains all types of media. Examples of such media include hard disks, magnetic media such as floppy disks and magnetic tape, optical recording media such as CDs and DVDs, floptical disks and magneto-optical media, ROM, RAM, Examples include hardware devices configured to store and execute program instructions, such as flash memory and the like.

Communication media include any information-carrying media that typically includes computer-readable instructions, data structures, program modules or other data in a modulated data signal, such as a carrier wave, or other transfer mechanism.

Examples of computer-readable instructions include not only machine language code, such as that produced by a compiler, but also high-level language code that can be executed by a computer using an interpreter or the like.

The above explanation is merely an illustrative explanation of the technical idea of the present invention, and a person having ordinary knowledge in the technical field to which the present invention pertains will understand that it does not depart from the essential characteristics of the present invention. Various modifications and variations are possible. Therefore, the embodiments disclosed in the present invention are for illustrating rather than limiting the technical idea of the present invention, and the scope of the technical idea of the present invention is not limited by such embodiments. It's not something you can do. The protection scope of the present invention shall be interpreted in accordance with the following claims, and all technical ideas within the scope equivalent thereto shall be construed as falling within the scope of rights of the present invention.

10...Rectangle 100...Ground 200...Target image 201, 202, 203, 204...Divided image 300, 400...Focal plane 1000...Image correction device 1010...Input section 1020 ... Image conversion section 1030 ... Output section

Claims (12)

A computer-implemented image correction method, comprising:
providing a first image;
providing viewpoint information for viewing the first image;
dividing the first image into two or more divided images along the lateral direction of the first image;
providing a second image converted from the first image by converting each of the two or more divided images based on the viewpoint information and the vertical length of each of the divided images;
Providing a second image transformed from the first image comprises:
determining a field of view length for each divided image based on the viewpoint information and the vertical length of each divided image;
determining a transformation length for each segmented image from the field of view length ;
The step of determining a transformation length for each divided image from the field of view length includes:
For each divided image, so that the ratio of the transformed length of the i-th divided image to the total transformed length is the same as the ratio of the field of view length of the n-i+1st divided image to the total visual field length. An image correction method comprising the step of determining a transform length, where n is the total number of segmented images. A computer-implemented image correction method, comprising:
providing a first image;
providing viewpoint information for viewing the first image;
dividing the first image into two or more divided images along the lateral direction of the first image;
providing a second image converted from the first image by converting each of the two or more divided images based on the viewpoint information and the vertical length of each of the divided images;
Providing a second image transformed from the first image comprises:
determining a field of view length for each divided image based on the viewpoint information and the vertical length of each divided image;
determining a transformation length for each segmented image from the field of view length ;
The step of calculating the transformation length for each divided image from the field of view length is performed using the following transformation formula;
the step of determining the transform length y _i of the i-th divided image using
Here, h _i is the field of view length of the i-th divided image, n is the total number of divided images, and T is the entire length of the target image. A computer-implemented image correction method, comprising:
providing a first image;
providing viewpoint information for viewing the first image;
dividing the first image into two or more divided images along the lateral direction of the first image;
providing a second image converted from the first image by converting each of the two or more divided images based on the viewpoint information and the vertical length of each of the divided images;
Providing a second image transformed from the first image comprises:
determining a field of view length for each divided image based on the viewpoint information and the vertical length of each of the divided images;
determining a transformation length for each segmented image from the field of view length ;
The step of determining the field of view length includes:
The step of calculating the visual field length using the visual field length calculation formula, the formula for calculating the visual field length h _i for the i-th divided image is
It is
Here, S is the line-of-sight length from the viewpoint to the focal point, θ _i is the viewing angle at the end of the i-th divided image on the opposite side of the line-of-sight, and θ ₀ is the viewing angle at the end of the first divided image on the side of the line-of-sight. An image correction method comprising the step of determining a field of view length using a field of view length calculation formula, where θ f is a field of view angle at a focal point _. A computer-implemented image correction method, comprising:
providing a first image;
providing viewpoint information for viewing the first image;
dividing the first image into two or more divided images along the lateral direction of the first image;
providing a second image converted from the first image by converting each of the two or more divided images based on the viewpoint information and the vertical length of each of the divided images;
Providing a second image transformed from the first image comprises:
determining a field of view length for each divided image based on the viewpoint information and the vertical length of each divided image;
determining a transformation length for each segmented image from the field of view length;
The step of determining the field of view length includes:
The step of calculating the visual field length using a visual field length calculation formula, the formula for calculating the visual field length hi for the i-th divided image is
and
Here, R is the line-of-sight length from the viewpoint to the end of the first image on the viewpoint side, θi is the viewing angle at the end opposite to the viewpoint of the i-th divided image, and θ0 is the line-of-sight of the first divided image. An image correction method comprising the step of determining a visual field length, which is a visual angle at a side edge, using a visual field length calculation formula. The image correction method according to claim 1, wherein the viewpoint information includes information that allows the position of the viewpoint to be specified with respect to the first image. The viewpoint information is
a viewpoint height (H) from the plane where the first image is located to the viewpoint;
a viewpoint distance (L) from a point where the viewpoint is projected onto a plane where the first image is located to a viewpoint side end (A) of the first image;
a viewing angle (θ ₀ ) at the viewpoint side end (A) of the first image;
a viewing angle (θ ₄ ) at the end (E) opposite to the viewpoint of the first image;
a line-of-sight length from the viewpoint to the viewpoint side end (A) of the first image;
a line-of-sight length from the viewpoint to an end (E) on the opposite side of the viewpoint of the first image;
Viewing angle of the first image (∠AOE)
The image correction method according to claim 5 , comprising two or more of the following. A computer-readable memory comprising computer-readable instructions, the computer-readable memory comprising:
Computer-readable memory, wherein the instructions, when executed on a computer, cause the computer to perform the method according to any one of claims 1 to 4. A computer-implemented image correction method, comprising:
providing a first image;
providing viewpoint information for viewing the first image;
dividing the first image into two or more divided images along the lateral direction of the first image;
providing a second image converted from the first image by converting each of the two or more divided images based on the viewpoint information and the vertical length of each of the divided images;
Providing a second image transformed from the first image comprises:
determining a viewpoint angle for each divided image based on the viewpoint information and the vertical length of each of the divided images;
determining a transformation length for each divided image from the viewing angle ;
The step of determining a transformation length for each divided image from the viewpoint angle includes:
The following conversion formula;
the step of determining the transform length y _i of the i-th divided image using
Here, H is the viewpoint height from the plane where the first image is located to the viewpoint, and L is the viewpoint side end of the first image from the point where the viewpoint is projected onto the plane where the first image is located. The image correction method is the viewpoint distance up to. A computer-implemented image correction method, comprising:
providing a first image;
providing viewpoint information for viewing the first image;
dividing the first image into two or more divided images along the lateral direction of the first image;
providing a second image converted from the first image by converting each of the two or more divided images based on the viewpoint information and the vertical length of each of the divided images;
Providing a second image transformed from the first image comprises:
determining a viewpoint angle for each divided image based on the viewpoint information and the vertical length of each of the divided images;
determining a transformation length for each divided image from the viewing angle ;
The step of determining the viewpoint angle for each divided image includes:
The following conversion formula;
the step of determining the viewpoint angle α _i for the i-th divided image using
Here, T is the overall vertical length of the first image, n is the total number of divided images, a i is the _vertical length of the i-th divided image, θ ₀ is the viewing angle at the line-of-sight side end of the first divided image, θ An image correction method in which _n is the viewing angle at the end opposite to the line of sight of the n-th divided image. The image correction method according to claim 8, wherein the viewpoint information includes information that allows the position of the viewpoint to be specified with respect to the first image. The viewpoint information is
a viewpoint height (H) from the plane where the first image is located to the viewpoint;
a viewpoint distance (L) from a point where the viewpoint is projected onto a plane where the first image is located to a viewpoint side end (A) of the first image;
a viewing angle (θ ₀ ) at the viewpoint side end (A) of the first image;
a viewing angle (θ ₄ ) at the end (E) opposite to the viewpoint of the first image;
a line-of-sight length from the viewpoint to the viewpoint side end (A) of the first image;
a line-of-sight length from the viewpoint to an end (E) on the opposite side of the viewpoint of the first image;
Viewing angle of the first image (∠AOE)
The image correction method according to claim 10 , comprising two or more of the following. A computer-readable memory comprising computer-readable instructions, the computer-readable memory comprising:
Computer-readable memory, wherein the instructions, when executed on a computer, cause the computer to perform the method according to any one of claims 8-9.